When the Apple M2 came out, it was tantamount to announcing that the M1 was about to become obsolete. However, upon closer inspection of its benchmark results, it seems that the M2 is just a slightly improved version of the M1. Although M2 and M1 both have incredibly powerful CPU capabilities, the overall evaluation seems to be only supported by the CPU and the amazing energy efficiency ratio. There is a very interesting point in the data recently released by Passmark. In the CPU Mark, M2 can kill everyone, and it is evenly divided with the core i9 12900K. However, in the Passmark comprehensive score, the M2 score is not high, and it is slightly behind the Core i3 12300. Such a disparity in evaluation is a bit surprising. Could it be that Passmark made a mistake somewhere and there is something wrong with the current scoring algorithm?
Intel was in disgrace when it launched the M1 family on Apple Silicon. After all, they were originally the dominant player in personal computers and only competed with AMD in terms of processor performance. Unexpectedly, Apple suddenly overtook them in a corner and slapped the two major manufacturers hard with the Apple M1. This result indirectly led to Intel’s rise in the 12th generation processor architecture, using a new large and small core hybrid architecture to bring very high performance and excellent energy consumption ratio.
In terms of performance alone, Intel still has amazingly powerful server processors, but the Apple M1 not only hit a new high in performance, but also has an unprecedented energy consumption ratio. This advantage continued with the advent of Apple M2 and is difficult to shake. Even if the progress of Apple M2 is criticized, there is probably not much controversy in claiming the title of the most energy-efficient processor in the field of personal computers.
During the Apple M1 era, various experts have shown their talents and exploited the performance scores thoroughly. This is also the case after the release of M2. Recently, the single-core CPU Mark score of M2 was available on the Passmark database. According to website screenshots, the M2’s CPU Mark score reached 4116 points, which is similar to the Core i9-12900’s 4116 points and the Core i9-12900KF’s 4209 points. It shows that in terms of single-core performance of the processor, there is no dispute that M2 is on an equal footing with these big players.
Passmark, meanwhile, has another CPU Mark score, which is determined by the average of the scores in the library. The more samples, the closer to the true performance. Different from the single-core CPU Mark score just now, the score here is the score of full-core operation. At this point, the score performance has undergone interesting changes. The Apple M2’s score of 14,933 points is slightly higher than the M1’s 14,577 points, but well below the Core i9-12900’s 36,418 points. The closest score to the M2 in the Intel camp is the Core i3-12300, which scored 14,955 points.
In other words, in Passmark’s score algorithm, M2 not only tied the Core i9-12900 in single core, but also tied with the Core i3-12300 in the comprehensive score. In other words, the M2 with only 8 cores (8 threads) has about the same overall Passmark performance score as the Core i3-12300 with only 4 cores (8 threads).
Such a judgment result may be unacceptable to some players, and they may even question whether Passmark favors Intel too much. Because similar problems also occurred when the company just launched the 10th version of the benchmark software in 2020. The controversy at the time was that a revised version directly caused AMD processors to fall wildly off the rankings, allowing Intel processors to be at the top, but Passmark did not immediately change the algorithm because of the doubts. The current version of the software is also version 10, with ARM and x86 versions respectively launched. This means that it can be executed natively on the M2 platform without recompiling. The score can directly reflect the performance. However, the results of the M2 and i3 are almost the same. It is obvious that the new algorithm this time is different from the projects seen in the general mainstream benchmark software.
Comparing M2 and Core i3-12300, the positioning of the two processors is very different. One is a newcomer with ultra-low power consumption and amazing energy consumption ratio, while the other is a desire to provide the best cost-effectiveness for budget users on desktop computers. Both sides have good advantages. Although Passmark is positive in the single-core test, it is obviously not optimistic about the multi-core performance of the M2, which has the same number of threads as the i3.
Would you refer to Passmark’s benchmark scores and consider Apple M1/M2 as only a 12th generation i3-level processor? The author believes that the running score results are for reference only, but in fact, it is also necessary to integrate the opinions of some players who are familiar with Apple products in order to have a more comprehensive understanding of Apple’s new products. If you just look at the running scores, you may ignore other advantages and disadvantages.
Source: KOCPC Chinese